Human Action Classification And Analysis

Monitoring suspicious activities in various places is important and has become a major research area. Suspicious activities if monitored properly then we can mitigate or avoid many unwanted incidents. In this paper we have proposed a system to detect human suspicious activity while operating an automated teller machine. Computer vision techniques are used to detect suspicious behavior and generate an alarm when required this in turn will help the security personnel which will help in improving the response time required to stop any unwanted incident.